South African Species of PeripcitidcB. 333 



not connected by a number of intermediate forms with the other 

 genera. 



Opposed to these forms of Perip atop sis, and differing from them 

 in several more or less important characters, is another South 

 African species, for which I have created a separate genus, described 

 below as Opisthopatus. 



Both these genera differ from the American forms on the one hand, 

 and from the Australasian on the other in several characters, such as 

 the position of the genital opening (between the legs of the penultimate 

 pair in the American and of the last pair in the Cape and Austra- 

 lasian forms), and the presence (in Australian and American), or 

 absence (in the Cape forms) of receptacula seininis in the female, 

 which are surely characters of generic importance. The number of 

 spiuous pads and the arrangement of the papillae on the feet are no 

 doubt of less value, but as these characters go hand in hand with 

 more important ones they are by no means to be disregarded. 



Several new species have recently been described by Bouvier 

 (Peripatiis tholloni from Gaboon, W. Africa) and A. Willey 

 (Peripatiis iiovce-britcmmcB from the island of New Britain), which 

 do not appear to fit into any of Pocock's genera. In that case, of 

 course, new ones would have to be made for their reception.* 



Specific Characters. The characters available for distinguishing 

 between the species of Peripatopsis are unfortunately very limited 

 in number, almost the only ones being the number of the legs, the 

 structure of the terminal rudimentary pair alongside of the genital 

 opening, and the shape of the papillae. 



The number of the legs is of value apparently in all species in 

 which there are not more than nineteen praagenital claw-bearing 

 pairs, besides the terminal rudimentary pair at least, no case of 

 variation has come under my observation. On the other hand, if 

 the number of the praegenital legs exceeds nineteen pairs, it is often 

 no longer constant in the same species, and its value as a specific 

 character becomes greatly diminished. 



A very important character is the presence or absence of claws on 

 the rudimentary genital legs, and I have used this to divide the 

 genus into two main groups. The presence or absence of a rudi- 

 mentary foot in the clawless forms will probably be found of specific 

 value, although I am not yet quite sure of the constancy of this 



* While this paper was in the press, A. Willey's researches on P. nouai-britannice 

 (in : Zoolog. Results based on material from New Britain, &c., collected by A. 

 Willey, Pt. I., 1898) came to hand. This author recognises Pocock's new names as 

 sub-genera, and creates a new sub-genus, Paraperipatus, for P. nov(e-britanniie. 



